social and ecological impacts of logging in · pdf filesocial and ecological impacts of...

31
I , r L_. SILVICULTURE AND ECONOMICS OF IMPROVED NATURAL FOREST MANAGEMENT (ITTO PD 4198) FORESTRY RESEARCH INSTITUTE OF GHANA SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA CONSULTANCY REPORT No. I COVERING THE PERIOD ," SEPTEMBER 2000 To 3, " AUGUST 2001 I^ By UNIVERSITY OF ABERDEEN M. A. Pinard and M. D. Swaine with K. A. Adam February 2003

Upload: dinhtuyen

Post on 26-Mar-2018

223 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

I ,

r

L_.

SILVICULTURE AND ECONOMICS OF IMPROVED NATURAL FOREST MANAGEMENT

(ITTO PD 4198)

FORESTRY RESEARCH INSTITUTE OF GHANA

SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA

CONSULTANCY REPORT No. I

COVERING THE PERIOD ," SEPTEMBER 2000 To 3, " AUGUST 2001

I^

By

UNIVERSITY OF ABERDEEN

M. A. Pinard and M. D. Swaine with K. A. Adam

February 2003

Page 2: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

I

ACRONYMS

Diameter at breast height

Food and Agriculture Organization of the United Nations

Forestry Department

Forest Inventory ProjectF1P

GHAFOSIM Ghana Forest Simulation ; a model developed to assist natural forestmanagement in Ghana

Ghana Timber Millers' Organization

International Institute for Environment and Development

International Tropical Timber Organization

Minimum Felling Diameter

National Redemption Council Decree

Research Working Cycle

Timber Export Development Board

dbh

FAO

FD

GTMO

11ED

ITTO

MFD

NRCD

RWC

TEDB

-

2

Page 3: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

Terms of Reference

Ecological Impacts of LoggingSocial Impacts of Logging

Abstract

1.0 Understanding the Social Implications of Logging1.1 Introduction

12 Background

2.0 Methodology2.1 Literature review of forest harvesting in the tropics2.2 Check list for field surveys

2.21 General Land-use characteristics of the Study Area22.2 Primary Stakeholders in the Forestry Sector2.2.3 Socio-economic characteristics of Forest Fringe Communities2.2.4 Population Growih2.2.5 Labour

2.2,6 Household Economy2.2.71ncome Distribution

2.2,8 Ethnicity2.2.9 Social Services

2.3.0 Traditional Administration2.3. , District Administration

2.3.2 Community Common Rights2.3,3 Rights to Water Resources2.3.4 Rights to Fishing2.3.5 Rights to Hunting2.3.6 Community dependence on commercial timber species2.3,7 Rights to gather Non-Timber Forest Products (NTFPs)2.3.8 Communal Accessibility to Forest Resources2.3.91ncreased Collaboration in Forest Management

TABLE OF CONTENTS

3.0 Concluding remarks

4.0 Ecological Impacts of Logging4.1 Introduction

4.2 Review of felling controls4.2. , Introduction4.2.2 Results and Discussion

a. Minimum Felling Diameter (MFD)b. Felling Cycle Lengthsc. Determination and selection of permissible cut treesd. Exploitation Intensity

4.2.3 Conclusion

5

5

6

6

67

7

7

7

7

7

7

77

7

7

8

88

8

8

8

89

9

9

If

It

1111

12

12

12

13

13

16

3

Page 4: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

I I-

4.31mpacts of regulating logging through MFD4.3. , Introduction

4.3.2 Objectives of the study4.3.3 Method of study4.3.4 Results and Discussion

a. Felling intensity and area disturbedb. Residual tree damage

4.4 Wood Loss due to Decay associated with Logging Wounds4.4. , Introduction

4.4.2 Methodology4.4.3 Results and Discussion

a. Unhealed logging wounds among timber speciesb. Stern size and logging wound decayc. Wood loss from logging wounds

4.4.4 Conclusion

5.0 References

AppendicesIA. Forest Reserves investigated for tree selection and removals

L IB. Table of Minimum Felling Diameter prescribed for timber (Class I & 2)species during different periods of the exploitation history in Ghana2. Study site and Methodology for Logging study

11 3. Methodology for study on wood loss due to logging woundsTables

I I. Minimum-felling diameter ranges used in Ghana during specific periodsbetween 1907-, 989

, 2. Comparative numbers of trees selected for felling and actual numbersL removed for some major species per compartment (128 ha)

FiguresL I . Histograms showing number of trees selected for harvesting (PR) and actual

number of trees removed (RM) per compartment (I28 ha) during logging inthe respective Forest Reserves at various periods of their exploitation history

. 2. Ground areas of plots disturbed by felling and skidding operations in twofelling intensity and stem selection treatments

3. Correlation between forest area disturbed during tree felling and skidding andnumbers of trees damaged

4. Box plot for DBH distribution of trees killed by felling and skidding operationsin sample plots at Pra-An urn Logging study site

5. Number of sterns of different dbh classes damaged or killed during tree fellingin two felling intensity treatments

L 6. Box plot showing decay depths measured in wounds assumed due to loggingdamage on lower boles of trees (selected species) 5 years prior to sampling7. Box plot for decay depths in the unhealed wounds among stern size (dbh)

classes of all sampled species8. Histogram of % basal area of stems in the small and large trees sizes of the

15 species affected with unhealed wounds and likely to be lost due to decay

16

16

16

16

17

17

18

20

2021

21

21

22

2223

24

27

27

28

30

12

15

14

17

18

19

20

21

22

23

4

Page 5: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

Ecological impacts of logging. Help develop methodologies to

> determine forest recovery through time after logging;> determine the number of seed trees to be retained after logging for effective natural regeneration;> review minimum felling diameter limit for sustainable timber harvesting;> determine wood loss in standing trees due to logging wounds and identification of economicalIy

feasible methods of log preservation in forests. Provide direction and support to project staff in data analysis and reporting. Provide support to the local project staff in the development of protocols for seed tree retention and yield

allocation

Social impacts of logging. Help provide an adequate understanding of the social implications of logging to project staff. Help develop methodologies to assess the impact of logging on the various usergroups of forests. Propose initigative actions necessary to restore the forest and maintain the community dependence on

forest resources

Terms of Reference

ABSTRACTThe report is in two parts, the first dealing with social aspects of the project, the second with ecological andsilvicultural issues

The social implications of logging in Ghana's forest reserves can be considered at both macro- and micro-levels. This project will focus on the micro-level impacts, particularly examining factors relevant to forestfringe communities, An examination of the SOCio-political context in which people living near forests isimportant for understanding how goods and services from forests support rural livelihoods and how forestryinteracts with agriculture. Institutional arrangements governing access to resources and benefit-sharingagreements include both traditional and district government institutions. Conflicts have arisen over access toforest products because of lack of consultation in policy development and incompatibility of regulationsrequiring permitting and social understanding of rights of access

We report on the project objectives to review application of minimum felling diameter limit in selective loggingand assessing wood loss in standing trees due to logging wounds in Ghana. We do this first by reviewing themechanisms used in regulating selective logging practices in Ghana from 1950 to 1990. Secondly weexamine the ecological impacts of regulating selective logging through minimum felling diameter in a loggingexperiment. The third part of this report deals with assessment of wood loss in trees damaged by logging.The results of each study are discussed in relation to long-term impacts associated with minimum fellingdiameter and felling damage. This report does riot include recommendations for the revision of the selectivelogging system in Ghana

1.0 UNDERSTANDING THE SOCIAL IMPLICATIONS OF LOGGING1.1 INTRODUCTION

Logging in Ghana's forest reserves has both macro- and micro-level impacts on society. At the macro-tevel,activities in the reserves affect society through their contribution to the national economy, through thedevelopment of infrastructure (physical and institutional), and industrial processing capacity. Also, becauselogging can affect the ecological functions of forests, and contribute directly or indirectly to change in forestcover, they impact the environmental services that forests provide to society. At a micro-level, loggingactivities may impact people living near forests in a variety of ways. For example, in Ghana many ruralpeople, particularly women, rely on non-timber forest products collected from the countryside, including theforest reserves (Falconer 1992). Logging activities within the reserves, and associated impacts adjacent tothe reserves, may influence access to nori-timber forest products and the quality and quantity of theseproducts available in the landscape. Logging activities, with the associated influx of machinery and workersfrom outwith the area, may bring conflicts, for example in terms of disturbance (e. g. , noise), risks to localpeople's safety. Alternately, logging activities may bring short-term positive impacts such as temporaryemployment, improvements to road networks, and increases in economic activity (e. g. , trade).

Attempts to manage Ghana's forest resource have varied over time as social, economic and environmentalchanges have occurred in the country. These management practices have shown an evolutionary trend fromthe establishment or reservation phase, to the development of silvicultural and management systems.Although the forests are managed for many purposes logging is the most conspicuous activity in the forest

5

Page 6: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

L.

^,

It is clear that logging PTOvides an important source of revenue in Ghana yet the socio-economic impacts oflogging, particularly at the micro"level, are poorly known.

The aim of this component of the project is to contribute to the understanding of the social impacts of loggingin forest reserves in Ghana. This aim is addressed through the following objectives: I) to examine theissues through a review of policy, project experience and Environmental Impact Assessments; 2) todetermine the impacts of logging on various usergroups, particularly from the perspective of the forestservice, timber contractors and local communities; 3) to develop methodologies to assess the impact oflogging on various usergroups; 4) to identify potential conflicts between timber management and theprovision of goods and services valued by people living near forests; and, 5) to propose initigative actionsnecessary to restore forest values that have been compromised, and to maintain the provision of goods andservices to people living near the forest resources. In the first year report, the focus will be on the firstobjective

1.2 BACKGROUND

Ghana's reserve forests (Appendix I) were demarcated for management as permanent forest estates for thepreservation of soil and water resources. conservation of biological diversity and sustainable production fordomestic and commercial use. Production reserves constitute almost 45% of the total forest estate, and fromthese forests the bulk of the nations timber is produced

The unreserved forest area is currently estimated by the Forestry Department to be 0.5 million ha, Theseareas were largely forested at the turn of the century but are rapidly disappearing as the forestland is givingway to other land uses. The unreserved forest patches lie within a matrix of, bush fallows and activeagricultural areas

The intact forests, mostly located in the Western Region of the country, have substantial canopy cover withcomposition similar to the reserved areas. The bush fallows are secondary growth from abandoned farmswith high potential for maturing into high forest if adequately managed. The agricultural lands arecharacterised by a relatively high density of trees on farms. Results from a forest inventory conducted by theForestry Department of Ghana with the support of the Overseas Development Administration (ODA), UK.revealed that off-reserve forest areas have about 268 million in of standing tree volume of suitable form tobe classified as timber. About 101 million in of the total volume is in trees greater than the minimum fellinglimits (FD undated)

There has been a rapid loss of forest resources from both reserved and off-reserved areas probably becauseof increasing population pressure. At the turn of the last century, the forest zone of Ghana covered about34% of the total land area. However, by I 987 over 75% of the land area originally covered by forests hadbeen cleared (Forestry Department, 1987). This corresponds to an annual deforestation rate of 0.84% (684kin2). FAO (1988) estimated annual deforestation between 1981-1985 in Ghana to be 220 kin2. If thesefigures are reliable, they suggest that the deforestation rate is decreasing. This trend may be a reflection ofthe small forest area outside reserves that remains to be felled (Agyeman 1994).

The total volume of timber extracted in Ghana between 1986-1992 was 7.8 million in (7.5% of standingvolume) of round wood (11ED et a1.1993). Increasing population pressures and its resulting impact on forestdegradation has been one of the most important factors influencing the evolution of policies, silvicultura!techniques and forest management practices in Ghana

2.0 METHODOLOGY

2. , Literature review of forest harvesting in the tropicsA review of literature, project documents, and environmental impact assessments was conducted todetermine established knowledge in relation to social implications of logging. This review was supplementedby field visits to timber extraction areas in Ghana. Results from the review were used to guide the creation ofa checklist of factors of interest to consider during field surveys. The review and field visits also were used toinform the formulation of a questionnaire for use with various usergroups. A review of policies andlegislation that have bearing on the exploitation and management of timber resources was conducted. Thehistory of forest management in the country was also reviewed

2.2 Check list for field surveysThe following check list was developed based on literature and experiences gathered from EnvironmentalImpact Assessment studies of a number of natural resource management projects

r.

6

Page 7: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

2.2. I General Land-use Characteristics of the Study AreaLand-use in Ghana is the direct result of the interaction of many factors related to the country's physical andhuman geography. The land-use zones relate to vegetation formations and are closely related to climaticconditions. Identification of the major land-use categories and determination of their relative importance fornatural resources, including biodiversity, will inform project formulation

2.2.2 Primary Stakeholders in the Forestry SectorFrom the occupational characteristics, the nature of households and land-use patterns in the study area,people that depend on or benefit directly from the extraction or utilization of the forest resources must beidentified and their characteristics described

rL

2.2.3 Socio-economic Characteristics of Forest Fringe CommunitiesThe social and economic characteristics of the communities around a study area, provides an understandingof the extent of dependency of the various identified social groups (stakeholders) on the forest resources. Italso presents an opportunity to assess the impact of the increased harvesting of timber in general andspecific species in particular on the various social groups in the study area.

2.2.4 Population GrowthAccording to the 1984 population census report, the populations of many rural communities in the forestzone are growing at 3.4%, which outstrips the national average of 2.6%. A rapidly growing population hasimplications for the maintenance of natural resources in these areas. Knowledge of migration and emigrationpatterns are also crucial for understanding change in local demands on the forest.

L,

I'L. J

2.2.5 Labour

Rural Ghana has an agrarian economy with about 92% of adults being actively engaged in small-scaleagriculture. The present study could explore current opportunities for rural employment in the forestry sector,and the interest within forest fringe communities in gaining employment in forestry. The field visits suggestthat, in general, logging activities bring very little job creation to rural areas as contractors rely on their ownworkers and tend not to recruit locally

I~L

~\

2.2.6 Household EconomyMost rural households derive their monetary income through farming especially cultivation of cocoa and foodcrops. Nevertheless, forest products may play a significant role as a source of supplementary andsubsistence income. It is common across the tropics for forest products to serve as a safety net, providingincome during periods of unemployment or resources during periods of scarcity. It will be interesting todetermine to what extent this pattern holds true in Ghana.

2.2.71ncome Distribution

The average annual household income should be estimated and compared with the national average.Patterns of inequality in the study areas should be highlighted.

2.2.8 EthnicityThe migrant community normally has socio-economic characteristics quiet different from indigenouspopulation. The migrants may have limited access to land. Although they can gain access to land for farmingthrough various arrangements there are still limitations, that may force them to adopt forest basedlivelihoods. Such livelihoods may either be facilitated by commercial logging come into conflict with it.

I_

2.2.9 Social Services

Social Services are recognized as an important prerequisite for development. Services accessible to forestfringe communities must be determined. Equally important is the quality and source of such facilities,therefore it will be important to determine if any of them have come as a result of forest exploitation.

2.3.0 Traditional AdministrationTraditional forms of administration are the responsibility of the chiefs and elders. The chief, who is thetraditional head of the community, and his elders are responsible for the day to day administration of thecommunities, organization of communal labour, revenue raising, ensuring environmental cleanliness, and theimplementation and monitoring of self help projects among others' The Chiefs in the communities hold thelands in trust for the people in the area. Therefore they receive royalties accruing from timber that have oflate become source of conflict between many traditional authorities and their subjects

7

Page 8: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

With regards to legislative and executive functions, the Chiefs with the assistance of sub-chiefs pass andenforce laws that they think are in the interest of the development of the traditional area. In addition to this,they also enforce bans that are backed by traditional beliefs. It will be interesting to know how themanagement of timber royalties is affecting the relationship between traditional authorities and their subjectsas well as the ability to make and enforce laws in the communities.

2.3. , District AdministrationThe District Assemblies (DAS) are the pivot of administrative, planning, developmental decision-making andrating authorities at the district levels. Members of the DAS are the District Chief Executive (DCE), thegovernments appointed political head, one elected person from each electoral area in the district, themember(s) of Parliament and other nominated persons not exceeding 30% of the total membership of theAssembly appointed by the President in consultation with the traditional authorities and other Interest groupsin the district. Each DA has a Presiding Member (PM) who is also elected by the Assembly. For thepurposes of planning, each Assembly has established a Planning Authority called the District PlanningCoordinating Unit (DPCU). These units are made up of such professional staff that the District PlanningAuthorities in consultation with the National Development Planning Commission (NDPC) agree upon (LocalGovernment Act 1993, Act 462). The DAS collect all royalties including those from timber accruing to thedistrict administration. The perceptions and aspirations of the DA members on the current role of logging inlocal council development agenda need to be captured.

2.3.2 Community Common RightsCommunity accessibility to natural resources depends on the nature and extent of common rights that prevailin the area. An understanding of the common rights therefore will indicate the extent of dependency andaccessibility of natural resources to the people in the study area. The team should study the followingcommon Tights likely to found in the study area

2.3.3 Rights to Water ResourcesAll the people in the communities, irrespective of status should the rights to water from streams, rivers andponds except when it is the product of someone's personal effort or creation. In most rural areas, water fromstreams, rivers and ponds are considered as 'free gifts of nature' and constitute the main potable water fordomestic use. Local processing industries depend very much on water from these sources for theiroperation. Some other end-uses of rural water are irrigation and aquaculture. Forest reserves providewatershed to most of the streams and rivers and their logging will there impact on rural water availability.

2.3.4 Rights to fishingPeople living close to watercourses have the right to fish in them without trespass. River fishing is basically anon-commercial activity in the forest area. Fishes caught are mostly used to supplement domestic protein in-take and therefore constitute a very important activity in rural economies. The intensity of fishing operationsdepends on the seasonality of the rivers, which to a greater extent are influenced by the state of the forestTo this effect logging may affect people's ability to fish efficiently

2.3.5 Rights to hunting . .Wildlife laws require that hunting is done with official permits. Nevertheless, people still follow the traditionalways and hunt anywhere including forest reserves with impurity. Hunting for bush meat is both a commercialand non-commercial activity. Bush meat constitutes a very important protein in-take of most people in ruralareas. The hunting expedition depends on the state of the forest. Most hunters spoke to now hunt from theforest reserves since the off-reserves are degraded. The positive and negative impacts of logging on huntingcan be assessed through hunter interviews and discussions with local food sellers who mostly serve asclients to the hunters

r'

L_,

I'

L

I~'

I.

2.3.6 Community dependence on commercial timber speciesIn Ghana, the forest is an integral part of the rural economy providing subsistence goods and services aswell as trade (Falconer, I 991). The forests provided food, fodder, fuel, medicine, building materials,materials for various households items as well as intangible benefits such as cultural symbols and ritualartifacts before the central government took control over their management and user rights. This studyshould examine the foregone benefits to communities as a result of their inability to remove timber speciesfor traditional or domestic usage

r

L

I~~ 2.3.7 Rights to gather Nori-timber Forest Products (NTFPs)According to forest laws local communities have the right to collect non-timber forest products from forestreserves for domestic uses with permission from the Forest Service Division. The NTFPs are very importantfor household use and are mostly gathered, hunted or harvested from the outside reserves. However, with

8

Page 9: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

-.

r

I*

L

^,

L

the recent levels of land degradation, more NTFPs especially the traded ones like cane, chewstick andmedicinal plants are collected from the forest reserves. During one of our rounds some NTFP collectorscomplained that a number of their commodities have become unavailable due to logging in the forestreserves. It is also possible that for the "light demanding" NTFPs logging will enhance their abundance. Weshould determine the influence of logging on NTFPs availability, quality, and sustainability

2.3.8 Communal Accessibility to Forest ResourcesForest reserves are fully vested in the State through the Forest Ordinance of 1927. Although the ownershipof land did not alter at the time of reservation the traditional owners have no right of access to the trees orland in the reserve except on permit from the competent government authority (Forestry Department). Themanagement of trees, the right to own, plant, use and dispose of trees within the forest reserves is controlledby the state, through the Forest Protection Decree, 1974 (NRCD 243)

Under the working plans of all forest reserves, the following communal rights are usually admitted in forestreserves by permit

. Communal rights to hunting, fishing, collecting of fuelwood, snails, medicinal plants

. Farming rights to admitted or allowed farms. Admitted farms are portions of land inside the forestreserves which had been farmed at the time of reservation and allowed to continue as such. These lands donot constitute part of the reserves even though they are situated inside them. These communal rights havebeen the subject of several disputes between the Government Forest Service (Forestry Department) and thecommunities

L_

r'

Forest tenure within what used to be the protected forest (now forest reserves) is also controlled by the stateGovernment regulated the use of trees within these forests through its Protected Timber Lands Act of 1959,which prohibited farming in concession areas until exploitation had ended

Concessions did not interfere with the customary rights of hunting, trapping, collection of firewood or naturalproduce other than timber. The tenure on protected forests is less restrictive than that in the forest reservesand the emphasis is on the type/species and use of the individual treesAnother type of forest tenure is through the purchase of timber rights. This is controlled by the ConcessionsAct of 1962, which regulates the granting of timber felling and harvesting rights. This tenurial system isrestricted to farms and the Protected forest areas which have good stocking of timber trees. The traditionalland owners, in this case the various Stool heads, are paid compensation or royalties through the ForestryDepartment. The local community, however, have the right of access to other nori-timber trees and theirproduce in the concession areas outside forest reserves.

Forest reservation by communities was usually on communal land in the form of Sacred grooves. Noindividual has the right to plant, use and inherit trees and tree products in most of these traditional forests orgrooves. In some cases the "Okomfo" or fetish priest of the community who is the care-taker of these sacredgrooves is accorded limited rights to sell tree products, but not the land. No individual has the right todispose of such lands. Customary law does riot encourage the reservation of communal or family land asforest by individuals if another member of the community or family needs it to farm

.

L

2.3.91ncreased Collaboration in Forest Management.A lackadaisical attitude of communities to the protection of the forest resources coupled with the inability ofthe Forestry Department to properly manage the forest led to increasing illegal harvesting of timber, whichcontributes to deforestation and loss of biodiversity. However with increasing harvesting of timber, the needfor collaboration in forest management has become more apparent. A new concept of community forestmanagement committees to deal with illegal logging and other forms of forest abuses has emerged inGhana. Initial observations suggest that these committees are prevalent and active in timber rich areas.The piloting of collaborative forest management systems in some of the study areas has increasedcommunity access to forest resources especially the harvesting of nori-timber forest products through freecommunity access to NTFPs for domestic purposes. The role that logging plays in bringing differentstakeholders together for a common purpose is worth investigating.I~'

I3.0 CONCLUDING REMARKS

The review has provided insight into the possible socio-economic impact of logging on local forestusers/fringe communities. However, the issues are complex and variable across the study area. Thehistorical lack of dialogue between the forest stewards (forestry commission) and local populations hascontributed to the development of ignorance and apathy among forest fringe communities. Special effort

9

Page 10: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

L

must therefore be made at the beginning of the study to win the confidence and trust of the respondents, andto explore with them the potential importance of the study and also its limitation in terms of direct impact oftheir lives

r*

I~"L

10

Page 11: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

4.1 Introduction

Sustainable forest management is to ensure forest productivity through ecologicalIy sound practices (i. esilviculture and logging). In this respect the maintenance of good forest structure after logging is of majorconcern.

Forest structure is controlled by four main variables namely, (a) amount of growing stock (b) diameterdistribution within stand, (0) species composition, and (d) Cutting cycle lengths (e. g. Smith, 1962; Oliver andLarson, 1996). It is generally said that these variables or attributes for regulating uneven-aged forestsstructure are more of forest art than science (Leuschner, 4990). However, mathematical models andguidelines, describing and directing foresters how to achieve balanced structure in uneven"aged forests arebased on these variables (Leak at a1. , 1987). For instance, diameter distributions of individual speciesexamined on fairly homogeneous tracts are found to follow either negative exponential or rotated sigmoidpattern (Muller, I 982; Lorimer and Frelich, I 984). However, these population structures depend on speciesand site (Leak, I 996). These variables are usually applied in combination to achieve a balanced structure(Leuschner, 1990), especially during the initial cutting period or first felling cycle referred to as adjustmentperiod (Recknage1,1913; Dewight, 1965)

In tropical countries, the most common variable for controlling forest structure are minimum felling diameter(MFD) and felling cycle (Dykstra at a1. , 1996; Dawkins and Philip, 1998). Since diameter (at breast height =1.3m) is convenient to measure and closely correlates with stem volume, it has become a convenientvariable used in forestry. For instance, diameter class models are used to generate future diameterdistribution and product yield based on initial distribution and known growth rates (Gadow and Hui, I 999)The knowledge of the structure of diameter distribution in a stand is also considered very vital in the designof felling system and silvicultural treatments especially, as they vary greatly with forest type and status ofadvance growth regeneration (Dawkins and Philips, 1998). Stand structure, specifically diameter distributionand species composition, is thus an important variable for monitoring stand development after harvesting.

In Ghana, various mechanisms to control exploitation with the aim of maintaining good residual forest andensuring sustainable production of timber have been used since 1907. These selective logging controlsrelate to species categorization, Minimum felling Diameter (MFD), Felling cycle lengths, and annualallowable cuts (Ghartey, I 989; Francois, 1989; Boakye-Dapaah, I 990; An on, I 995; An in akwa, 1996),

Because sustainable forest management is seen as one way to reduce the loss of bio-diversity and forestdegradation, there is a pressing need to assess the consequences of these regulatory mechanisms on theforest structure and dynamics (Pelissier at a1. , 1998), especially when there is no post logging SIIviculturaltreatment

4.0 ECOLOGICAL IMPACTS OF LOGGING

It is believed that the negative impacts of logging on residual forest can be reduced through effectiveplanning, tree selection procedures, and good logging practices (Hendrison, 1990). A current approach toreducing negative effects of selective logging in the tropics is through Reduced Impact Logging (RIL)practices (Benault and SISt, I 997; SISt at a1. , 1998). RIL techniques are built on strict rules that guide theplanning and control of logging operations (Dykstra and Heririch, I 9961 Pinard at a1. , I 996). These includepre-logging forest inventories, topography survey and layout plans for skid trails. However, significantimpediments remain to the widespread application of RIL technologies in the tropics (Putz at a1. , 2000;Dykstra, 2001). Apart from the cost and training needs identified as general problems (Sarre and Efransjah,2001), successful application of RIL requires adequate input of specific local information such as how manytrees to be felled, which species to be harvested, and which trees are to be retained (Agyeman at a1. , I 999)

The effects of timber harvesting on forest health and ecological stability depend on logging intensity andprevailing physical, and biological conditions (Hendrison 1990). In the same manner, the future value of anypreviously logged forest for timber production and biodiversity conservation depends to a great extent onhow much damage was done during the previous harvesting (Gullison and Hardner 1993). On these baseswe review the past regulatory practices in the light of current forest conditions.

4.2 Review of felling controls4.2. , Introduction

To understand the effect of felling controls on forest exploitation and residual forest stand, seven forestreserves were selected for study (Appendix I). At least one forest in each of the five regional forestadministrations in the high forest zone (i. e. Ashanti, BTOng-Ahafo, Eastern, Central, and Western regions)was selected for study. For each selected Forest Reserve (FR) the past and present forest management

11

Page 12: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

~^.

I_,

plans (where available) were studied to find the control mechanisms applied since it was first logged,Controls searched for were:

. Minimum felling diameters (MFD) for various species

. Cutting cycle lengths

. Methods of selecting trees for felling and retention as seed trees or residual stand

In each FR, timber exploitation records for at least two compartments logged between the periods 1951-1960; 1961-1970; 4971-1980; and 1981-1990 were compiled. For each compartment, the number of treesprescribed for felling and the actual numbers of trees felled for a period were compiled from the compartmentregister so as to determine the prescribed and actual harvest intensity and relate these to the controlmechanisms during the periods.

4.2.2 Results and discussion

Forest management regulatory mechanisms applied between 1910 and 1990 included 4 versions ofminimum felling diameters, 3 different felling cycle lengths, and 3 different methods for selecting thepermissible number of trees to be felled per compartment. There have also been species groupings basedon silvicultural or commercial considerations. The classified economic species constitutes the exploitedgroup

(a) Minimum Felling Diameter (MFD)Application of felling diameter limits in Ghana commenced with the passing of Timber Protection Ordinancein I 907 that sought to protect the felling of immature trees (Taylor, 1960). Since then, there have beenrevisions of these felling limits in 1910, I 958.1 972 and I 989 (Ghartey, 1989) and recently in 1997 (Of OSu-Asiedu at a1. 1997). A comparative table of the felling diameter schedules for 1907-1989 is provided asAppendix (2) and a summary of the limits of the various schedules is given below (Table I)

Table I. Minimum-fell^hg diameter ranges used in Ghana during specific periodsbetween 7907-1989. Period refers to approximate revision dates of the LimitsThe Limits varied with species and period

Period

I907,949'I950,971b1972-1988c1989'

Felling limits between 50 cm and 110 cm dbh are being used in Ghana. The bases for fixing the previouslimits are not known. However, since I 989, they are supposed to indicate a point in the species diameterdistribution where the average national stocking shows a sharp decline. The latest revision also adopteddiameter ranges similar to that of 1989, but classified some species into lower or higher limits based onadditional variables such as average diameter increment of different DBH classes, ten-year average annualexploited volume, stocking km' above 40 cm dbh, and prevalence of decay in exploited sterns of respectivespecies (Of OSu-Asiedu at a1. , I 997). We observed that, since I 972, Ghana has adopted felling diameterlimits that are higher than in many countries in West-Central Africa. For example the Ghanaian MFD are 10-40 cm higher than that of Cameroon, 10-60 cm higher than in Cote d'Ivoire and :t 10 cm of that in Liberia(Panen and de Graaf, 1995; Dykstra et a1. , 1996)

Sources: a = Logan (1947); b = FD (1955); c ^ FD (4972); d = Francois (1989).

Minimum Felling Diameter(MFD) limits (cm. dbh)58.68,8770,90, I I O70, I I O50,70,90,110

(by Felling Cycle LengthsBetween I 950 and 1989 three different lengths of felling cycles were instituted in the management ofreserved forests in Ghana. A felling cycle of 25 years was started in the I 960s, when working plans for someforest reserves were first produced and the respective forests were opened for timber exploitation andsilvicultural treatment. In 1972, when the first cycle was only 12 years old, a management operation termed"salvage felling" was introduced to remove so-called over mature trees. It was explained that the then on-going enumeration surveys had shown a preponderance of large diameter trees that were dying and orlosing their timber value because of defects. It was believed necessary to remove them through commerciallogging before they were lost to natural mortality. Under this operation, all reserved forests with or withoutmanagement plans were to be logged within 15 years, The records indicate that the salvage felling, whichshould have ended in 1987, continued until 1990 in the absence of an alternative yield regulation methodSalvage felling offered an unprecedented chance for second cycle felling in some forest reserves earlier than

12

Page 13: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

expected. In 1990, a forty year felling cycle was introduced based on the average time of passage (Top)estimated as the time for most of the high value species to grow from the next lower diameter class to themature class (Boakye-Dapaah, 1990). This review does not discuss the rationale of the 40-year felling cycle,as its effect cannot be seen in the present stocking data.

(0) Determination and selection of permissible cut treesIn Ghana, two broad approaches have been used for determining permissible cut. They are the formula anddirect diameter methods. Between I 950 and 1971 three formulae methods developed by Jack, Kinloch, andKandambi, all colonial foresters were introduced (An on, 1962). They were all derivatives of the Brandismethod (OSmaston, 1968), used in the Teak forests of the Far East, and the classical European methods(Brasnett, 1953). The Jack, Kinloch and Kandambi formulae determined the permissible cut in basal arearecruitment (An on, , 962), but the cut was prescribed in number of sterns for each economic species perannual coupe. The yield was selected from sterns within the felling diameter classes and usually the biggersterns were chosen first. When the total calculated yield was not obtainable from the available stock ofexploitabte trees, the deficit could not be taken from the lower diameter class

Between I 972 and I 989 and alongside the salvage felling regime, only the Minimum Felling Diameter (MFD)was used as the means to determine the permissible cut. By this method all stems of a given speciesrecorded to be above the MFD during a 1007, timber Inventory in a compartment were selected for felling(An on, 1972; An on, I 995).

After the 1989 national forest inventory, another formula method was introduced. This formula is based onthe population of the sterns above the MFD and those in the next lower dbh class. It also seeks to providebetween 409". to 60% retention of trees above MFD for canopy structure, seed production and bio-diversityconservation (An on, 1995). The current form of the formula is given as

Z = 0.5Y + 0.2X - for the moist forest, andZ = 0.25Y + 02X - for the dry forest

where:

Z = total permissible number of stems,Y = the number of stems of a given species above the MFD, andX = the number of stern of a given species in the next lower diameter class to the MFD

The formula is applied at the compartment (128 ha. ) level and the Y and X estimates are obtained from a100% timber inventory of all commercial class I trees. Species that are exploited at dbh 270 cm areinventoried down to 50 cm dbh. Those exploited at 50 cm are inventoried down to 30 cm

According to Vanclay (1993) the yield prescribed by this formula, if considered against diameter recruitmentand stem mortality over a 40 year felling cycle, can be sustainable only when the ratio of X:Y is 3:1, asituation that is rare in the TMF. Adam (1999) also points out that the 40-60 percent retention envisaged bythe above formula cannot be attained, because the algebraic equation for the retention input was incorrectFor instance when the value of Y equals X, it means 70% and 45% of the mature commercial sterns will becut in a moist or dry forest, and the respective retention will become 30 and 55% instead of 40 and 609'.respectively. As the ratio of X to Y increases the retention value diminishes, but that is not what is intended.The continued use of this formula therefore makes sustained timber production doubtful.

A simulation model (GHAFOSIM) was developed by Alder (1989), but has not yet been used to determinethe permissible cut at forest reserve level. This is due to the limited number of species for which incrementdata were available at the time of developing the model. However, it has been used at the planning level toprovide estimates of stand mortality, species group time of passage, and national annual allowable cut, allvariables used as inputs for the current yield regulation formula (An on, 1995)

(d) Exploitation IntensityAlthough the forest reserves were available for exploitation in the late 1950s, full-scale entry appeared tohave occurred in the 1970s (Figure I) or at the onset of the Salvage felling regime. The ForestryDepartment records for the 7 Forest Reserves examined show a gradual increase in both the numbers oftrees selected for felling (i. e. prescribed yield) and the actual trees removed over the years (Figure I). Thelevels, however, vary with Forest Reserve. While prescriptions varied between 80 and 780 stems percompartment, actual removals were been between 25 to 320 sterns (i. e. less than 50% was exploited).

The number of stems prescribed for felling depends mainly on the list of commercial species of the period,species richness and the number of trees above the MFD in the compartments due for exploitation. Thus

19

Page 14: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

^,

there was the tendency to select more trees for exploitation when there were many stems above the MFD.Similarly, more sterns would be exploited if the selected yield was composed of many high value species,For instance, the selected trees at Atewa Forest were made up of mainly high value species (Meliaceae) andthey were all exploited despite the low number (Figure I). At Draw River Forest. one high-value species(Heritiera utilis), constituted over 40 % of selected trees, and because over 72% of this species wasexploited, it finally made up 54% of the total exploited number (Table 2). Again, because Heritiera isgregarious in this forest, there were many trees above the MFD (70 cm dbh), and an average of 156 stemswere selected per compartment, for exploitation, resulting in high felling intensity (Figure I). In the moistsemi-deciduous forest type where Tr^IOChiton soleroxylon also dominates, similar selection and removaltrends were observed. For instance, at Offin shelterbelt, Tano Suhien and Mamang Forests, Tr^IOChitonsoleroxylon constituted 47%, 26% and I8% of selected tree whiles it made up 43%, 31 % and 29% ofremovals in the respective forests (Table 2). The overreliance on MFD as a criterion for selecting trees forfelling results in high felling intensity, especially, when there is a higher stocking of sterns z: MFD. It wasobserved in this study that, the amount of trees removed, was influenced more by the concession holders'preference, than by the forest mangers' selection.

an

500

co 400

31 300d 200

co*,

Dawl^rer

-------. - --.---..

o

ICO

o

1951^0 1951-70 ,971-80 1981-go 1991-99

*

is

$;

5

I"

Year

^,!;

4co

o3

liana I^"'

co~

02oZ

t,

4:0

35/1

300co

E 231

:I 2.0. 150

~

o

a

=;.~

'95/60 193F70 Igy^O I^tSO Is^99

e.

a;^

IVbrrng Rrer

^

t;"*

a=

o

z ICO

50

o

^"

52

SI;\is

a*

niR

I '!a I^! ICl

sai

Y^

"

3:0

co 2^iEg 2.0

o ,50

z ICO

co

o

orris'eitch^t

800

700

600in

^ 50031 4009 300o

zan

100

o

1951^0 19^-70 '97/80 Igyi. go I^I. ^9

+.

W, war,

'55/8i 1951-70 1971e0 1581-go 1991.99

;;

DR^

in

Yea

,+

$1^^' Its e

jig

a*

1951-60 t95, -70 1971-80 1981-90 1991-99

DR?

' BRIn

s;*

Year

600

I .I^

' I D, MYea

an

E 400o

co 300o

6 200Z

^

^

SL, "'in

100

o

'95/6i 1951-70 1971-80 1981-90 1991-99

.FR

I DRvi

;Iaa

Year

14

Page 15: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

.

-,

600

500

= 400

co 300a,+"

o

o 200Z

Tano Suhien

Figure I. Histograms showing number of trees selected for harvesting (PR) and actual number of treesremoved (RM) per compartment (f 28 ha. ) during logging in the respective Forest Reserves at variousperiods of their exploitation history. Due to hilly terrain at Atewa Range, tree selectibn there, was based onyield marking of a few valuable species that were allremoved

100

o

,951-60 1961-70 1971-80 1981-go 1991.99

------ - - -. -,

;e

Table 2, Comparative numbers of trees selected for tolling and actual numbers removed for somemajorspecies per compartment (728 ha. ). Removal was 25-54% of total number selected

FR

;^;!SE

Year

.,

^

I Dl'R

I Dan

Draw

re I^s: I

Species Selected

Wawahi

Hertetiera urn^

Khaya IvorensisP^fadeniastrum aOther species (36)Total

Subim

Annans toxicaria

Tr^IOChiton sNesogordonia pOther species (52)Total

Selected Trees

NoTotal

Mamang

Tr4?IOChilon s.Entandrophragma uPerlcopsis elataOther species (32)Total

1562124186387

Tano

Suhien

Tn^ochiton sTerminalIa superbaEntandrophragma cOther species (30)Total

40.35.46.248.4100

V.

964551

367559

No.S ecies

Trees Ex 10ited

112311452209

Offin

Shelterbelt

Tn^oohiton s.PI:)to deniastrum aKhaya IvorensisOther species (40)Total

17.28.1

9.165.6

100

96

3521388

540

Note: High value species (Entandrophragma, Khaya & Perlcopsis) were always removed in preference toothers even when sterns allocated were fewer. At Draw River FR stems of Khaya removed were more thanallocated. Source: Data compiled from Compartment Registers of respective Forest Reserves at ForestDistrict offices in Ghana

71.8147.658.327.954.0

9'o

2123

12

93149

Tn^oohiton sEntandrophragma u.Khaya arithothecaOther species (25)Total

17.86.5

3971.9100

% Total

23

169

83131

53.614.86.724.9100

A further examination of the Forest Reserve felling records showed that, between I 960 and 1989, removalsin six of the seven sites, were below 200 trees per coin^artment (428 ha. ) which agrees with the thresholdlevel of 182 trees per compartment or 1.5 trees hectares' that has been suggested by Hawthorne (1993). Itwill be important to examine the influence of these felling in tensities on the residual stand against thebackground of reported forest degradation during these regimes of selective logging.

F'.

I I

L.

21.951.123.5

25

26.7

3124

1568138

17.612.26.963.4too

.

108

4513

2454/1

14.115.4

8.162.4

100

,

32668.6

71.4I7.5

25.5

17

83157

26.310.93.2

59.6100

137

1012

134293

22.5I7.4

10.949.2100

73.96.2588.937.343.5

46

1188146

46.834

4.1

45.7400

29.81.81454.4100

42.6

2.284.6

35.935.5

68

911

69157

31.5077.5

60.3too

49.690

9251.4

53.6

43

5.7743.9

100

15

Page 16: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

I '

L

L. .

4.2.3 Conclusion

Repeated cutting targeted mainly at larger trees or all mature trees of relatively few species, offers little or nochance for distributing tree removal in a balanced way across all size classes for individual species.Exploitation control mechanisms (most notably Minimum Felling Diameter, tree selection methods, andcutting cycle lengths) used in the previous management regimes in Ghana, can not be seen to have aimedto maintain a balanced species mix in the residual stands especially because removals were influenced bymarket demand

Cutting cycle lengths of 25 and I5 years were used mainly to remove larger diameter trees but riot totransform the diameter distribution into a balanced structure at the end of the second cycle in 1987. This isdemonstrated by the stand structure in 1989 (F1P, 1989). The yield formula introduced in 1990 seeks toaddress the diameter distribution by retaining a certain percentage of the exploitable trees, but the retentionpercent was not based on any prior ecological or silvicutIural studies. It is not clear whether the retention of40-60% stem numbers above the MFD will give adequate density of canopy trees to prevent creation of largegaps especially in the partly degraded and the dry forests. Minimum felling diameters are an importantaspect of the regulation mechanisms but the rationale is influenced more by economic, than ecological orsilviculturat considerations

4.31mpacts of regulating logging through MFD4.3. , Introduction

Injection of investment capital into the timber processing industry in 1983, as part of Ghana's EconomicRecovery programme (ERP) and trade liberalisation policies, has contributed to the expansion of processingcapacity, which TEDB (1996) estimated at 2.7 million in' per annum. The current annual cut prescribed bythe Forestry Department is one million m' (An in akwa, 1996), insufficient to meet the increased nationalprocessing capacity. Therefore, there is mounting pressure from the timber industry for increasing harvestintensity (GTM0,2001). The industry has strong bargaining power because of its immense contribution tothe general economy (Banahene, 2001). The government is thus under immense political pressure to assistthe timber industry as an important component of the national economy, as well as protect the forestresources as a national heritage. Forest resource managers are meeting these requests from industry andtheir obligations to the government with recommendations for the increased use of under exploited species

The viability and implications for increasing exploitation of additional species need to be investigated. Someexamples of questions that could be raised are as follows

Can the harvesting intensity be increased if more species are removed?How much loss in terms of stem stocking, species composition, and future timber yield can beexpected if felling intensity is increased and more species are harvested, under the current selectivelogging practice?Can the forests absorb more disturbances than they have already taken?

4.3.2 0byectiVes of StudyIt was assumed that large diameter trees are usually associated with large crowns and when felled, willcause more damage, by creating larger canopy gaps which favour regeneration of pioneer and non-pioneerlight demanding species. Alternatively, selection of more smaller diameter trees will cause less damage butcreate smaller canopy gaps which favour shade tolerant timber species. The research objectives were

I. To compare logging damage (ground area disturbed, stern damage, and canopy loss) in areas harvestedat felling intonsities equivalent to current annual allowable cut (26.3m' ha~') prescribed by Ghana forestservice & two times as much (52.6m' ha~') in combination with two stern size selection criteria (small andlarge diameter stems)2. To explore the implications of stern diameter as a criterion for tree selection to reduce logging damage.

4.3.3 Method of studyWe examined the ecological implications of increasing harvest intensity by testing two levels of fellingintonsities and tree selection criteria and comparing treatment effects on numbers and size of felling gaps,canopy loss. skid trail density, and proportion of trees damaged

These objectives were to test the hypothesis that logging damage (stern damage, ground disturbance, andcanopy loss) is positively related to felling intensity and size of tree felled. The study was conducted at Pre-

16

Page 17: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

L

,,. An urn Forest Reserve between January and September 2000. A description of study site and methodology

is provided in Appendix 3.

4.3.4 Results and Discussion

(a) Felling intensity and area disturbedOver an area of 32 hectares each, extraction of 1600 in' and 840 in' of timber (equivalent to 53m' and 26 in'ha~' respectively) resulted in 5.5 kin and 3.9 kin of skid trail network respectively. In terms of area, skid trailsin the high intensity site disturbed 2 ha (i. e. 12.5 in' in") and 1.3 ha (i. e. 15.5 in' in~') in the low intensityharvesting site. These areas disturbed by the skidder (07 crawler tractor), were equivalent to 6.25% and 4%of the area of the high and low intensity logging sites respectively. Vegetation was completely destroyed andsoils compacted or turned over by the tracks and logs dragged along the soil surface.

The skid-trail length increased by doubling the felling intensity, but the increase was only 40%. This mayindicate that proportionate Iy less damage will result to forest from skidding a given harvest volume if theharvest intensity is increased. In fact it is unlikely that the relationship between ground area disturbed andharvest intonsities are linear (Pantil and Gullison, 1998). Some logging simulation models also suggest thatdamage could be reduced further if skid trails are made linear. For instance Gullison and Hardner (1993)achieved a 2570 reduction in logging damage by requiring main roads and skid trail to be linear in asimulation model. Several studies also show that pre-planned skid trails do reduce logging damage (e. g.Johns at a1. , 1996; Badault and Sist, 1997). For instance Johns at a1. (1996) found that for a tree harvested,unplanned logging affected a ground area that was more than 100 in greater than planned operations.Jackson et a1. (2002) also reports 25% ground area disturbance in the form of skid trails, logging roads andlog landings in a Bolivian forest at a harvest intensity of 12.1 in' ha' and attributed the high level of groundarea disturbed to unnecessarily long skid trails.

The devastating effect of skidding was also partly due to the wide trail widths that were created as a result of frequentmanoeuvring of the crawler tractor. This usually occurred when the trail meandered and caused the log to swing overthe trail edges. In this study it became obvious that felling intensity and size of the log extracted had the effect ofincreasing the width of skid trail and consequently disturbing larger ground area. On average, the trails in the highintensity treatment sites were 0.9m wider than those in the low intensity site, and it was 0.5m wider in the sites with largertrees than that with smaller trees felled.

Although the sizes of the trees felled did riot show any significant effect on tree fell"gaps, it seemed to have influencedthe skid trail width and sizes of log-yards. In the high intensity treatment the site with large stems had a log yard that was50% larger than the one with smaller stems. In the low intensity treatment the difference was 26% greater for the largesterns. It thus appeared that larger log yards were needed in respect of larger trees, possibly for machine manoeuvre.

The harvested volumes gave a stem felling intensity of 5-6 trees ha' and 2-3 trees ha~' for the high and low in tensitiesrespectively. The results indicated significant differences in ground area disturbed by felling at low and high intonsities.Obviously areas disturbed through tree felling in high intensity harvest (median 242.3= in', N = 32) were larger thanthose in low intensity harvest (median =, 38.8m') sites.

However. felling gap areas created by felling large diameter trees (median =151.9 in', N = 32) appeared smaller thanthose created by felling many smaller trees (median = 217.1 , N =32). Statistical analysis, however, did riot show thedifferences to be significant. The results suggest that, overall, felling many smaller trees create gaps similar to fellingfewer large size trees. This is true especially as felling clusters of trees do result in creating large gaps. Cluster fellingwas more frequent in the study sites where many smaller trees were felled (Figure 2). However it may be cautioned thatthe results could have been biased due to the narrow range of diameter sizes selected for felling in the two tree sizetreatments.

1000

B

or

Eb"P

= Booin

,

a

Boo

012

rL

aa

.C=

et9

400

200

,..........

...........

...........

...,.......

.........,.

o

".

.. "...,...

32

Tree Size

Large sterns

32

...........

,..,..,....

1000

32

Operation

^Fel, rig

''11Skidahg

Smallslem.

N

E."D

^ 600D

co

32

800

ein

.==o.

o

A

GEE

400

200

02

o

... ". ". "..

.............

.............

".

.............

.......,..,..

32

Felling Intensity

Low

32

.............

.............

..,..........

.....,.......

........,....

"

Operation

High

a,

^Felling

^Skidding

17

Page 18: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

L

.

.

Figure 2. Ground areas (in') of plots disturbed by felling and skidding operations in two fell^hg intensity andstem selection treatments. Low tolling intensity = 26m' ha', High felling intensity = 53m he' . Large stems =fewer number but larger diameter stems, Small stems = more stems but smaller diameter.

As anticipated, the forest canopy loss due to logging was higher in the high intensity treatment (Mean loss =20.8% ^. 1.3 SE, n = 129) than in the low intensity treatment (Mean loss = 41.65% ,^ 1.2 SE, n = 107). Thiseffect is similar to the ground area disturbed for the reason that twice the felling intensity did not producetwice as much effect on canopy loss. However, the result does corroborate other studies showing that thereis a negative correlation between tree harvesting intensity and post logging forest canopy cover (Webb,I997; Whitman at a1. , 1997) in other words, there is a strong linear relationship between canopy loss andharvest volume (Pereira Jr at a1. , 2002)

(b) Residual tree damageThe study failed to reject the null hypothesis that there are no differences in the number of stems damaged(injured or killed) in a high and low felling intensity. Some studies have reported positive correlation betweenharvest intensity and proportion of residual stem damage (e. g. Webb, 1997; Sist at a1. , I 998). In this study itdid riot appear that the number of sterns damaged (i. e. injured or killed) during logging does necessarilyrelate to the volume harvested

r~

120

100

80

^coOiCo

cot>

co

co,,co.,-

o,-

coD

E=Z

E

60

E40

Ground area disturbed

Figure 3. Correlation between forest area disturbed during tree felling and skidding and numbers of treesdamaged

This result corroborates the observations of Johns at a1. (1996) who did not find any correlation between thenumber of harvested trees and the amount of damage. It was riot feasible to do a correlation analysisbetween felling intensity and numbers of damaged stems as the investigation was dealing with variationbetween only two intonsities. However, it was observed that between the two felling intonsities, it was thearea disturbed that had more influence on the number of trees damaged (Figure 3)

The area disturbed through tree felling varied significantly between the two harvest in tensities and washigher in the high intensity treatment sites. This indicates a positive association between felling intensity andarea disturbed during tree felling and consequently numbers of trees damaged. However, as discussedabove, doubling the harvest intensity did not double the area of skid trail, for example, nor the number ofdamaged stems. Therefore, it can be expected that increased harvest intensity may riot increase sterndamage significantly

The overall effect of felled tree size on number of stems damaged in felling gaps was also not significant inthis study (N = 256, F = 3.56, df = I, P = 0.06). In single-tree felling-gaps the number of stems damaged arereported to show positive correlation with the dbh of felled trees (e. g. Jackson at a1. , 2002). However, over a

18

a

20

a

r

,' PC...-

^,,.Bd,.,,=. A. a"a O' a

.a'..

a

.,

o

",D

- a

.."

D'

a*

B

o

a

J

a. a-

a

.

,F

a

r'*

.

.

200

.

P

a

.

400

,^.

600

a

.

800

Logging operation

' Skidding

. Felling1000

Page 19: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

larger area, as in this study, the association may become weaker due to overlapping of felling gaps andresultant reduction in average number of stems damaged per tree felled. For instance, Panfil and Gullison(1998) observed in a logging experiment in Bolivia that as harvest intensity increased, trees tended to fall atleast partially on existing gaps so that per tree damage declined. They concluded that the area of vegetationdamaged by logging was a quadratic increasing function of harvest intensity

For the two logging operations examined (tree felling and log skidding), skidding appeared the moredestructive (Figure 4). This was because many live trees had to be removed, to clear a trail in order to reacheach felled tree. The natural instinct for the skidder operator to avoid large trees and waterlogged terrainalso increased the skid trail length, resulting in more fatal ity, especially to smaller diameter stems that thecrawler tractor could easily drive over. Skidding thus resulted in killing more smaller size trees (i. e. theadvance regeneration = 5- 19 cm dbh) than felling in the two treatments. Comparatively, felling operationsalso killed more medium diameter trees (> 20-49 cm dbh) than skidding (Figure 4). This was mainly becausefellers lack skills in directional felling and could not effective Iy prevent felling onto nearby large trees. Theseobservations support the recommendations that if unnecessary skid trails can be avoided, for instance byplanning trail layouts, and logging undertaken when soils are dry, and if tree fellers are trained to applydirectional felling, logging damage to residual trees could be greatly reduced (Jackson at a1. , 2002).,

L_

ra

A

92in

>.D

^;"an

co

co

DN.

Figure 4. Box-plot for DBH distribution of trees killed by tolling (A)and skidding (8) operations in sample plots at Pra-An urn Logging study site,

No significant differences were observed in the proportion of stand stem stocking damaged in the two fellingintonsities. However, the proportions varied with diameter class (Figure 5). In the high intensity site thepercentage damage tielling+ skidding) ranged from 9% ^ 2(SE) to 20% = 4 (N = 32) and in the low intensity itranged from 11% ^ 2(SE) to 16% t 3 (N = 32) with the higher dbh class (50+ cm) always recording the least.This means that distribution of damage across dbh classes was similar in the two felling intonsities and theproneness to logging damage was more dependent on the size of the residual stems. Similar trends arereported from many logging studies in the tropics (e. g. Readhead, 1960; Mensah, 1966; Benault and Sist,1997). In forests of Indonesia, Bertault and Sist (1997) observed 40% damage to residual sterns (> 10 cmdbh) under higher felling in tensities (> 80 in' han ) and when trees of diameter > 50 cm dbh were felled.

The management implication here is that the influence of logging on the residual forest diameter distributionwill vary depending on the stand diameter composition at the time of logging and the severity of damage.Therefore, it would be unwise to prescribe high harvest intensity for example, if there was also a high densityof small individuals of commercial species that would be killed during extraction.

Damaged stern dbh class

"

59 un

a ,.

to19un 20-490n 50+cm

40

"

a 30=

bb\co

>D

8 20

B

,e

in

Eon̂

to10

oN=

Damaged stern dbh class

64

5.9 cm

r

64

10-19cm 20-49cm 50+cm

~.

Be 64

19

Page 20: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

50

^co01to

Eto^

coEcoco~ 20

.

coDE

^ 10

40

A

30

,-.

Felling intensity

Figure 5. Number of stems of different dbh classes damaged (14) or killed (8) during tree fell^hg in two folling(Low -26m' ha~' and High -53 in' ha") intensity treatments. DBH class I= 5-9 cm, ' class 2 = 1049 cm; class

3= 20-49 cm, ' class 4 = 50 + cm) in 32 sample plots per treatment

B

4.4 Wood Loss Due to Decay Associated with Logging Wounds4.4. , Introduction

Dying economic timber trees or live hollow trees constitute a major source of timber production loss whenthey are felled and have to be abandoned because of wood degradation or hollowness resulting from decay.Another disadvantage of harvesting hollow trees is the extent of damage as compared with log volumerecovery. For some species and stern sizes, the volume obtainable will riot be worth the residual damage leftbehind (Imayabir at a1. , 2000).

Injuries to standing trees, especially wounds extending beneath the sapwood zone, render trees morevulnerable to pathogen infestation that may lead later to decay (Strouts and Winter, I 994; Greif andArchibold, 2000; Whitford, 2002). Other micro-organisms, arthropods and birds may further break down thedecaying wood so that finally a cavity or hollow is formed (Strouts and Winter, 1994) CUIminating in loss ofcommercial volume. Decay caused by fungi make stems of trees less commercially useful and in somecases completely useless for manufacture into lumber or veneer (Taintsr and Baker, 1996). Nicholson(1958) drew attention to logging damage such as bark and wood injuries in the tropical forests that favour theentry of pathogens which are likely to reduce tree survival rate or cause bole distortion making the timbercommercially unsuitable (cf Sist and Nguyen-The, 2002). Logging wounds also account for high mortalityrate immediately after logging. For instance. Sist and Nguyen-The (2002) found a significantly higher standannual mortality of 2.6% in logged forest as against 4.5% in unlogged Dipterocarp forest of East Kalimantanwithin 4-5 years after logging. They rioted that the increased mortality in the logged forest was mainlyamongst injured trees

Most logging damage studies provide quantitative or qualitative assessment of residual tree damageincluding percentage of stems or basal area of trees killed or injured (e. g. ITTO project P079/90). Forexample, in the logging damage study conducted at Pra-An urn, between 8 and if % of sterns > 5 cm dbhwere estimated killed and 5-6 % left injured and standing. But post-logging conditions of wounded treeshave riot been studied in managed tropical forests in order to estimate potential commercial wood loss andto develop measures for preventing wood degradation and timber loss. Techniques to identify degradation instanding trees in order to exclude them from production and prevent loss of production time are limited in thetropics. Most loggers rely on the experience of tree fellers who either use the "knock on wood" method or 90sawing with chainsaw into a suspected wound to identify presence of hollows and cavity size in trees (Bell,1971; Trockenbrodt et a1. , 2002). These methods were found unreliable for preventing harvesting of hollowtrees in Sabah, Malaysia (Trockenbrodt at a1. , 2002)

o

N. 32 32 32 32

Low

.... "...

stem dbh class

32 32 32 32

"

a

I;^a

ER

High

2

3

4

I~'

20

Page 21: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

The extent and severity of logging damage is highly dependent on harvest intensity and mode of conductingskidding operations. This study aims to provide additional information to support the implementation ofappropriate felling and skidding practices in order to improve residual tree conditions

The objectives of this study wereTo assess the incidence of healed and unhealed loggings wounds among different timber species and

different stern sizes in compartment RWC1-3 of Pra-An urn forest five years after logging damage.To assess the incidence of wood decay and discoloration in unhealed wounds in order to estimate the

potential loss of commercial timber due to decay in association with mechanical damage to trees duringlogging

4.4.2 MethodologyThe study was conducted through field observation of wound conditions on wounded trees andmeasurement of wound surface area and decay depth at Pra-An urn forest. Full methodology is provided inAppendix 4.

44.3 Results and Discussion

(a) Unhealed logging wounds among timber speciesA total of 390 wounds were found on the lower bole of 359 trees of 15 selected timber species incompartments RWC, -3 of Pra-An urn forest. Amongst the wounds encountered, 264 were completelycovered with regenerated bark and were assumed healed. The other 126 had no regenerated bark and outof these 103 had decay. Most trees had single wounds but 31 had two wounds each in different conditionThree tree species Gelt^^, Turfeanthus and Storeulia dominate the sample because they dominate currentstock of timber species at the site

The results showed that the distribution of the healed and unheated wounds among the sample trees wasdependent on species. Species with low numbers of unhealed wounds were those that appeared capable ofrapid bark regeneration after wounding. The converse may be true for those with higher occurrence ofwound decay. This agrees with other observations (Harmon et a1. , (1986) and Spies at a1. , (1988) cf. Franklinat a1. (2002) that wood decay following wounding is related to species.

400

350

EE

300

=

Q.co^DC.o

250

200

^150

Tree species

Figure 6. Box plot showing decay depths measured in wounds assumed due to logging damage incurred onlower bobs of trees (selected species) 5 years prior to sampffng. Anr =An ingeria sp, Ant=Antiaris toxicaria,

Gem = Gelt^S mildbraedi7zenkeri, GP = Gelba pentandra, Ea =Entandrophragma an901ense, Go=Guareacedrata, Ki=Khaya Ivorensis, Nes=Nesogodomia papaverifera, PIO =PI, todeniastrum amcanurn,

Ptm=Pterygota macrocarpa, Pyc = Pycnanthus an901ensis, Str= Storeulia chinopetala, 711 =Tr^IOChitonsoleroxylon, Ts =Terminalia superba, Tur= Turfeanthus Africana

Regenerated bark is essential for preventing further infestation (Philips and Burdenken, 1992) and forrestoration of cainbial activity around the full circumference of the bole (Thomas at a1. , I 995). The anatomyand chemistry of the bark tissues of the sampled species, which are riot well known, may explain part of the

21

100

50

o gAnr

Ant

Cem

GP

Ea

^

a;

a=a ^

Gc

IC

Nes

pipPtm

PycSIr

Tm

Ts

Tur

Page 22: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

variation. Field description of the tree bark showed that all the species that had over 70% healed woundswere those with thick, fleshy, soft or firm bark (e. g. Entandrophragma an 901ense, Gelba pentandra,Turreanthus amcan a, and Storeuffa rhinopetala). Decay in wounds (Figure 6) occurred mostly in species withbrittle, thin or dry and fibrous bark (e. g. Pycnanthus an901ensis, Geltis, Guarea cedrata and Pterygotamacrocaipa). Variations in incidence of wood decay after wounding may also be explained by the naturaldurability of the sapwood and heartwood of the respective species as rioted in several studies (e. gIfebueme, 1977 of Onuorah, 2000; Supriana, 1998; Guariguta and Gilbert, 1996). For instance, mean wounddepths were higher in species such as Pycnanthus an901ensis, Storeulia rhinopetala and Anti^ris toxicariathat have less durable timber (Irvine, 1961 ).

(b) Stem size and logging wound decayBelow 50 cm dbh. decay depths appeared to increase with increasing stem diameter (Figure 7). Meandepths were 31 mm :t8 SE, 61 mm :t, 2 SE and 112 min ,:30 SE in the 2029,30-39 and 40-49 cm dbhclasses respectively. However, it fell to 47 min ^10 SE in the 50+ cm dbh class.

According to A1bers and Heyd, (2000) the ability of the wounded stem to recover quickly from injury isinfluenced by several factors including growth rate, which is also related stern dbh (or age), and crownstatus. For instance, Whitford and William (2002) found that incidence of hollows in Eucalyptus treesincreased with tree size, and larger hollows were more likely in highly series cent crowns.

Stern diameter increments for Ghanaian trees reported by Alder (1990) showed that in all diameter sizeclasses, trees with shaded crowns have lower increment than those in sunlight. For example, stem incrementfor trees 10-70 cm dbh growing wholly or partially under main canopy is seen to rise rapidly between 10-30cm dbh and drops after 30 cm dbh. However, trees with overhead light show a decreasing growth rate withincreasing diameter. Under these circumstances, if rate of stern increment determines the rate of woundhealing, then depending on the crown status, results of wound healing after logging damage can be highlyvariable. Thus the variation in decay depths found among stem sizes in this study do not necessarilyindicate a definite relationship between decay depth and stem size

500

400

^

a 300cobDC=o

^

^200

100

o

DID,

N=

Figure 7, ' Box plot for decay depths in the unhealed wounds among stern size (dbh) classes of all sampledspecies

Stem size (dbh) class

20-29cm

24

(c) Wood loss from logging woundsAs the sample was riot selected per unit area, results are discussed only in relation to the total sample. Theresults showed that wounds sustained during logging by trees in 11 of the sampled species, led to decaywithin 4-5 years after logging. Larger wounds size also appeared to facilitate development of decay

*eu$

31

20.39an

04

017

40.49cm

20 51

50+am

22

Page 23: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

90

80^co

70^S^ 60IC

50tUco.

40t,

co 30"

20Co

^R 10

o

,S^' ^S* <;S' C;^ <:9' G' q;" ^, 9" <<' ,^" ,-^" ^;" <." A*" <.<:.Tree Species

Figure 8. Histogram of % basal area of stems in the small (20-49 cm dbh) and large (;a 50 cm dbh) treessizes of the I5 species affected with unhealed wounds and likely to be lost due to decay, Anr =Aningeria sp,

Ant=Antiaris toxicaria, Gem=Cellis mildbraedi7zenkeri, GP=Gema pentandra, Ea=Entandrophragmaan901ense, Go=Guarea cedrata, Ki=Khaya Ivorensis, Nes=Nesogodomia papaveriit?re, PIO=Plotsdeniastrum

atribanun, Ptm=Prefygota macrocama, Pyc=Pycnanthus an901ensis, Str=Sterculia rhinopetala,Tri=Tr^IOChilon soleroxylon, Ts =TerminalIa superba, Tur= Turfeanthus Amcana

About 26m' (34 %) of the total basal area (77 in') of wounded trees sampled had unhealed wounds, Thesewounds were in various degrees of deterioration indicated by decay depths ranging between I to 350 min(Figure 6). The percentage basal area loss also ranged from 29.4 to 35.2% among the 4 diameter classes,The quantity of timber likely to be lost as a result of logging-induced wounds will therefore be influencedsignificantty by the species composition of injured stems.

The majority of stems that receive fatal wounds during logging are those in the smaller diameter class < 50cm dbh. For example, it is estimated from the Pra-An urn logging study that between 10 and 20 stems ha' (8-25%) in the <50 cm dbh and less than 2 stems ha" (4-14%) of the >50cm dbh, sustain injuries but are notkilled. The smaller size class (< 50 cm dbh) are sterns below the exploitable diameter limits applied for therespective species and may thus not be legally possible to salvage them. This implies a complete loss tocurrent and future production. Although few larger diameter stems are inflicted with fatal wounds, because oftheir large diameters, they constitute a loss of (2-6%) of pre-logging stand basal area ha' (DBH >50 cm) atPre-An urn if the wounds lead to decay

Wound decay after logging also contributes to wood loss because infested wounds provide weak locationson the bole that facilitates stems breakage during storms. For instance. Mattheck (1995) found, in 800hollow trees, that those with less than 30% of their radius intact were broken

4.4.4 Conclusion

About 30% of logging wounds found among the species examined had developed into decay. Basal injurieslike those found in this study are known to be a major factor in the incidence of bark rot and heart rot inhardwoods (Nair and Sumardi, 2000) for which there is no control other than reducing the incidence ofwounds

Wound treatment in selectively logged forest may not be feasible for economic and operational reasons. Forinstance, in intensive tree care practices no wound treatment has yet been found which protects largewounds against decay organisms for a usefully long period (Strouts and Winter, I 994). In reviewingtreatment of pruning wounds, Mercer (1979) concluded that though many treatments may encourageregenerated bark production around the wound, they do little to prevent infection by decay fungi (cf. Philipsand Burdekin, 1992)

This study raises questions for the need to study wound-healing mechanisms such as the anatomical. andphysiological basis for susceptibility or resistance to decay after wounding in tropical timber trees, andmeasures to control stern damage during logging. This is very important for the fact that five species that

23

. 2049cm . 50+cm

r..

Page 24: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

showed high tendency for decaying (Pycnanthus an901ensis, Antiatris toxicaria, Pteiygota macrocarpa,Geltis innbraedi, "zenkeri and Tr^IOChiton soleroxylon) constitutes nearly 40% of current standing volume oftimber trees in Ghana (Ghartey, 1989)

REFERENCES

ADAM, A. R. , OFOSU-AslEDU, A. , DEI-AMOAH, C. , and AsANTE-ASIAMAH, A. , 1993. Wood waste andlogging damage at Asukese and Afram Headwaters forest reserves. Better utilization of tropical timberresources in order to improve sustainabitity and reduce negative ecological impact. In report of ITTOProject P079/90, FORIG, Kumasi, Ghana

ADAM, A. R. , 1999. Timber yields determination and allocation in selective logging system. In Ghana Journalof Forestry v01. (8) I 999 p23-30

AGYEMAN, V. K. , ABU-JUAM, M. , and HAWTHORNE, W. D. , 1999. Towards better forest HarvestingSummary of findings of the research project Impact of harvesting on forest mortality and regeneratibnin the hi^h forest zone of Ghana Forestry Research Institute of Ghana, Kumasi. , 8pp.

ALBERS, J. and HEYD, B, 2000: Incredible Trees How They Grow and Survive. Michigan State University,upper peninsular forestry extension paper 8/2000.

ALDER, D. I 990: GHAFOSIM: A prediction system for natural forest growth and yield in Ghana. Final

report for Ministry of lands and Natural Resources, Department of Forestry, (Under assignment fromODA, UK). I 14pp

ANINAKWA, B. 1996. The derivation of the annual allowable cut Paper presented at the 1996 Generalmeeting of the Ghana Institute of Professional Forestsrs

ANON, 1962. Forestry Department annual report. 1962. ACcra, GhanaANON, ,972. Ghana Forestry Department Annual Report. AGCraANON, 1995. Timber yields from the forest reserves of Ghana. An analysis of the implications of sustainable

forest management. Forestry Department (Planning Branch). Kumasi, Ghana. Conducted by theGhana Forestry Commission, ACcra (unpublished)

BANAHENE, M. F. , 2001. The changing face of Ghana's timber trade. In The ATBT Newsletter V01.13(2001). P 27.

BEAUDET, M. , and MESSIER, C. , 2002. Variation in canopy openness and light transmission followingselection cutting in northern hardwood stands: an assessment based on hemispherical photographs.Agricultural and Forest Meteorology. 110 (2002). P 217-228

BELL, T. I. 1971: Management of the Trinidad Mora forest with special reference to the Matura ForestReserve. Government printers Trinidad 70pp

BERTAULT, J-G. and SIST, P. , I 997. An experimental comparison of different harvest intonsities withreduced-impact and conventional logging in East Kalimantan, Indonesia. Forest Ecology andManagement. 94 (1997) p 209-218

BOAKYE-DAPAAH, A, s, K. I 990. Overviews of interim yield regulation measures. Paper presented atForestry DepartmentI Timber Contractors Workshop. Kumasi, Ghana. 13-14 November I990,

DAWKINS, H. C. and PHILIPS, Ms. 1998. Tropical moist forest silviculture and management: a history ofsuccess and failure. CAB International 359pp

DEWIGHT, T. W. , 1965. Confusion worse confounded; the sad story of forest regulation. For. Chron. 41 (1)p76-83

DYKSTRA, D. P. , 2001. The old and new of reduced impactlogging. Topical forest Update. V01. ,2 (2), 2001p 3-4.

DYKSTRA, D. P. , KOWERO, G. S. , OFOSU-AslEDU, A. and KIO, P, (eds. ). 1996. Promoting Stewardship offorests in the humid forest zone of Anglophone West and Central Africa. UNEP/CIFOR collaborativeresearch project final report. GIFOR, Indonesia , 03 pp

DYKSTRA, D. , and HEINRICH, R. , 4996. FAO Model code of Forest Harvesting Practice. FAO, Rome, 85ppFERGUSON, S. H. and ARCHIBALD, D. J. 2002: The % power law in forest management; how to grow dead

trees. Forest Ecology and Management I 69 (2002) p 283-292FORESTRY DEPARTMENT, 1995. Manual of Procedures, High Forest Management, Section -D, Stock

survey and yield allocation. Planning Branch, Forestry Department, Kumasi, Ghana.FRANCOIS, J. H. 1989. Minimum Felling Limits. Chief Conservator of Forests, office circular, 7th November

I989, ACcraFRANKLIN, J. F. , SPIES, T. A. , PELT, R. V. , CAREY, A. B. . THORNBURGH, DA. , D. R. , LINDERNMAYER,

D. B. , HARMON, ME. , KEETON, W. S. , SHAW, DC. , BIBLE, K. and CHEN, J. 2002: Disturbances andstructural development of natural forest ecosystem with silvicultural implications, using Douglas-firforests as an example. Forest Ecology and Management 155 (2002) p 399-423.

GADOW, K. V. , and HUI, G. , 1999. Modelling forest development. KluwerAcademic Publishers. 213 pp

24

Page 25: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

-. J

.^

^\

GHARTEY, K. K. F. 1989. Results of the Inventory Project. In Wong, J. L. G (ed. ) Ghana Forest InventoryProject Seminar proceedings 29-30 March 1989, ACcra. Forestry Department of Ghana, Kumasi. p 32-46.

GREIF, GE. and ARCHIBOLD, 0. W. 2000: Standing-dead tree component of the boreal forest in centralSaskatchewan. Forest Ecology and Management. I 31 (2000) p 37-46.

GTM0. , 2001. Fate of the Ghana Timber Industry. Position paper submitted to the parliamentary selectcommittee on lands and forestry in protest of Timber Concession allocation. GTMO, Kumasi.

GUARIGUTA, M. R. , and GILBERT, G. , 1996. Interspecific variation in rates of trunk wound closure in aPanamanian lowland forest. Biotropica 28 (, ). P 2329.

GULLISON, R. E. , and HARDNER, J. J. , 1993. The effects of road design and harvest intensity on forestdamage caused by selective logging: Empirical results and a simulation model from the BOSqueChimanes, Bolivia, Forest Ecology and management. 59 (1993). P I-, 14.

HALL, J. B. , and SWAINE, MD. 1984. Distribution and ecology of vascular plants in tropical rain forestvegetation: Forest vegetation in Ghana. Junk The Hague

HARMON, M. H. , FRAKNLIN, J. F. , SWANSON, F. J. , 1986: Ecology of coarse woody debris in temperateecosystem. Adv. ECol. Res, 15 (1986) p I 33-302

harvesting guidelines for a reduced-impact logging project in Malaysia.HAWTHORNE, W. D. 1993. Regeneration after logging. Findings of a study in Bia-South Game production

forest reserve in Ghana. 0. DA Forestry Series No. 3 ODNNRl, Chatham.HAWTHORNE, W. D. and ABU-JUAM, A. 1995. Forest protection in Ghana. IUCN, Forest Conservation

Series No. ,4.1UCN, GlandHENDRISON, J. 1990. Damage-Controlled logging in managed Rain Forest in Sunname. Ecology and

management of tropical forests in Sunname. Agricultural Univ. Wageningen. The Netherlands 204pp.HENDRISON, J. , 1984. Harvesting Systems. In project LH/UVS 01, Human interference in tropical rain forest

ecosystems, Annual report 1983. GELOS, Paremaribo, SurinamIFEBUEME, S. C. , 1977. Natural durability of wood with particular reference to West African Timbers. In

Proceedings of the international workshop on wood preservation held at the Forestry ResearchInstitute of Nigeria, Ibadan, 7-, 2 November I977, p 41-96

IMIYABIR, Z. , TROCKENBRODT, M. and JOSUE J. 2000: An Assessment of Hollow Logs and OtherLogging Residues from Deramakot Forest Reserve, Sabah, Malaysia. Paper presented at GroupSession 3,0500 'Forest Operations in The Tropics' of XXI IUFRO World Congress 7-, 2 August 2000,Kuala Lumpur Malaysia. 7pp.

IRVINE, F. R. 1961. Woody Plants of Ghana -with special reference to their uses. Oxford Univ. Press. 868ppJACKSON, S. M. , FREDERICKSEN, T. S. , and MALCOM, J. R. , 2002. Area disturbed and residua stand

damage following logging in a Bolivian tropical forest. Forest Ecology and Management. 168 (2002) p27,283

JOHNS, J. S. , BARRETO, P. , and UHL, C. , 1996. Logging damage during planned and unplanned loggingoperations in the East Amazon. Forest Ecology and Management 89 (1996) p 59-77

KINNEAR, P. R. , and GRAY, CD. , 2000: SPSS for windows made simple Release 40 Department ofPsychology University of Aberdeen. Psychology Press UK 380pp

LEUSCHNER, W. A. , 1990. Forest regulation, Harvest scheduling, and Planning techniques. John Wiley &Sons, Inc. 281 pp

LINDERMAYER, DB. , CUNNINGHAM. R. B. , POPE, M. L. , GIBBSON, P. and DONNELLY, C. F. 2000: Cavitysizes and types in Australian EUCalypts from wet and dry forests-a simple rule of thump for estimatingsize and number of cavities. Forest Ecology and Management 137 (2000) PI39-, 50.

LORIMER, CG, and FRELICH. I. E. , 1984. A simulation of equilibrium diameter distribution of sugar maple(Acersaccharum). Bull. Torrey Bot. Club (111) pi93-199.

MATTHECK, C. , 1995, Biomechanical optimum in woody sterns. In Gartner B. L. (Ed). Plant sterns:Physiology and functional morphology. Academic Press Sari Diego. P 75-90

MENSAH, K. 0. A. 1966 An analysis of logging damage in compartment 4 of Tano, Suhien forest reserve,Ghana, Ghana journal of science v01.6 nos. I&2 p 63-69

MERGER, PC. 1979: Attitudes to pruning wounds In Aboric J, 3 (1979) p 457-465NAIR, K. S. S. , and SUMAR01,2000. Insect pest and diseases of major plantation species. In K. S, S. Nair

(ed) Insect pests and diseases in Indonesian forests GIFOR Bogor Indonesia. PI5-3NICHOLSON, D. I. , 1958: An analysis of logging damage in tropical rain forest, North Borneo. Malaysia For.

21, P235-245.ONUORAH, E. 0. , 2000. The wood preservative potential of heartwood extracts of Mireia excelsa and

Erythroph/eum suave Qins. Bioresource Technology 75 (2000) p I 71-173PANFIL, S. N, , and GULLISON, R. E. , 1998. Short"term impacts of experimental Experimental timber

harvest intensity on forest structure and composition in Chimenes forest Bolivia. Forest Ecology andManagement I 02 (1998) p 235-234

.

,

25

Page 26: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

L

L

PELISSER, R. , PASCAL, J. P. , HOULLIER, F. , and LABORDE, H. 1998. Impact of selective logging on thedynamics of a low elevation dense moist evergreen forest in Western Ghats (South India). In ForestEcology and Management 105(, 998) PIO7-1/9.

PEREIRA Jr, R. , ZWEEDE, J. , AsNER, G. P. , and KELLER, M. , 2002. Forest canopy damage and recoveryin reduced-impact and conventional logging in Eastern Para, Brazil. Forest Ecology and Management168 (2002) p 77-89

PHILIPS, D. H. and BURDENKIN, D. A. 1992: Diseases of Forest and Ornamental trees. Second Edition.Macmillan Press LTD. Honk Kong 581pp

PINARD, M. A. , PUTZ, F. E. , TAY, J. , and SULIVAN, T. E. , 1996. Creating Timber harvest guidelines for areduced-impactlogging projectin Malaysia Journal of Forestry 93 (to) p 41-45

PUTZ, F. E. , DYKSTRA, D. P. , and HEINRICH, R. , 2000. Why poor logging practices persist in the tropicsConservation Biology. 14 (2000). p 951-956

READHEAD, J. F. , 1960. Analysis of logging damage in lowland rain forest, Western Nigeria, NigeriaForestry Information Bulletin (New Series) No. 10. Federal Government Printer, Lagos.

RECKNAGEL, A. B. , 1913. The theory and practice of working plans. 1st Edition. John Wiley and SonsPP

SARRE, A. , and EFRANSJAH, . 2001. Catching on to reduced impact logging. Tropical Forest Update. VolI I (2). PI "2.

SIST, P. and NGUYEN-THF, N. 2002: Logging damage and subsequent dynamics of a dipterocarp forest inEast Kalimantan (, 990-, 996). Forest Ecology and Management I 65 (2002) p85-103.

SIST, P. , NOLAN, T. , BERTAULT, J-G. and DYKSTRA, D. , 1998. Harvesting intensity versus sustainabilityin Indonesia. I 998. Forest Ecology and Management. 108,998) p 251-260.

SIST, P. , SHEIL, D. , 1<ARTAWINATA, K. , and PRIYADl, H. , 2002. Reduced-in pact logging in IndonesiaBorneo: Some results confirming the need for New silvicultural prescriptions. Forest Ecology andManagement (Article in In press). 6439 (2002) p 1.3

STROUTS, R. G. and WINTER, T. G. 1995: Diagnosis of ill health in trees. Department of Environment,Forestry Commission. UK. 307pp

SUPRIANA, N. , 1988. Studies on natural durability of tropical timbers to termite attack. InternationalBiodegradation V01.24 (4-5) p 337-341

TAINTER, F. H. and BAKER, F. A. 1996: Principles of forest pathology. John Willey& Sons, INC. NY. 805ppTEDB. , 1996. Annual Trade Statistics. Timber Export Development Board. Takoradi. GhanaTHOMAS, V. , PREMAKUMARl, D. , REGHU, C. P. , and PANIKAR, A. 0. N. , 1995. Anatomical and

Histochemical Aspects of Bark Regeneration in Hevea brasilliensis. Annals of Botany 75 (1995) p 421-426

TROCKENBRODT, M, , IMIYABIR, Z. and JOSUE, J. 2002: Hollow logs and logging residues fromDeramakot Forest Reserve, Sabah, Malaysia. Forest Ecology and Management 165 (2002) pi41-150

WEBB, E. L. I 997. Canopy removal and residual stand damage during controlled selective logging inlowiand swamp forest of northeast Costa Rica. Forest Ecology and Management. 95 (1997) p 117-129

WHITFORD, K. R. 2002: Hollows in ianah (Eucalyptus marginata) and marri (Gorymbia calophylla) trees I,Hollow sizes, tree attributes and ages. Forest Ecology and Management 160 (2002) p 20,214.

WHITFORD, K. R. and WILLIAMS, MR. 2001: Survival of jarrah (Eucalyptus marginata Sin. ) and marri(Coolinbia calophylla Lindt. ) Habitat trees retained after logging. Forest Ecology and Management 146(2001) p 181-, 97.

WHITFORD, K. R. and WILLIAMS, MR. 2002: Hollows in jarrah (Eucalyptus marginata) and marri (Gorymbiacalophylla) trees 11. Selecting trees to retain for hollow dependent fauna. Forest Ecology andManagement 160 (2002) p 215-232

WHITMAN, A. A. , BROKAW, N. V. L. , and HAGAN, J. M. , 1997. Forest damage caused by selection loggingof mahogany (Swientenia macrophylla) in northern Belize. Forest Ecology and management 92 (1997)p 87-96.

.

rL,

'*

235

26

Page 27: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

APPENDIX, :

IA: Forest Reserves investigated for tree selection and removals. Table gives year of reservation, forestarea, forest type, location, and first logging date for the FRs. Forest type is after Hall & Swaine (, 981):MSNW = Moist semi-deciduous Northwest type, MssE = Moist semi-deciduous southeast type, WE = WetEver reen, and UE = U land Ever reen.

Forest Reserve

(FR)

Atewa Ran e

Wawahi

Tano Suhien

Draw River

Maman

APPENDICES

Offin-Shelterbelt

Note: Exploitation did not start at same time in all FR due to several reasons (e. g. Accessibility, alternativetimber sources available to Concessionaire, nearness to processing mill or log market) as reflected in theduration between date of reservation and 1st exploitation.

IB: Table of Minimum Felling Diameter (MFD) prescribed for timber (Class I & 2) species during differentperiods of the exploitation history in Ghana. Non-Classified (uric) species were generally exploited at 70 cmdbh or at the discretion of the Forest utilization officer. SP = Species with low stocking or restricted ran eand are to be exploited under special permit from the forestry commission. R = Species that had their MFDrevised a ain after the 1989 schedule

Subim

Year ofReservation

1926

1929

1934

AreaKm

1937

1938

1951

2323

1956

138.9

Forest

Type

84.4

235.4

Afzelia africana

54.4

UE

Airetia belle

60.3

MssE

S p e cie s

A1bizia ferru inea

238.3

MSNW

Location

A1bizia z

WE

AIStonia boonei

Lat

MssE

Ainp him a s p te roca rpioide s

610 N 036W

MSNW

An ingeria s p

540N I06W

MSNW

620N 228W

Lon

nO

Antiaris toxicaria

91a

512N 220W

A n tro ca ryo n in ICra s ter

616N 103W

1st

ExploitationEnt

Berlinia s

rL

xis klaineana

640N 203W

Bombax buono ozense

647N 247W

Ca n a riuin Soh wein furthii

1967

Gelba pentandra

1907-50a

1963

--,

Geltis spp

1962

Chrysophyllum spp

uric

1986

uric

1978

Co aifera salikunda

uric

1957

Cordia spp

uric

1971

Periods And MFD

Cylico dis cus ga b u n e n SIS

1950-71b

uric

Cynometra anata

uric

I~*

Daniella s

uric

uric

^.

Dial^^in aubrevil/ei

87

uric

010S

uric

uric

^,

1972-88c

En ta n droph ragm aan 901e n s e

uric

uric

uric

uric

70unc

En ta n dro h ra gin a ca n dollei

uric

uric

ros sariza-minika

70unc

En ta n drop h ra gin acylin dric urn

uric

uric

70 unc

>, 989d

uric

uric

,

70 unc

Entandro hragma utile

68

110

70unc

Er thro hleum sp

70

uric

uric

70unc

70

uric

70unc

70

uric

uric

70unc

70

uric

uric

110

87

110

uric

70unc

90

uric

uric

70unc

110(90R

uric

uric

70unc

70

uric

70unc

70

uric

uric

uric

50

87

uric

70unc

70

uric

70

uric

uric

70 unc

70

87

uric

uric

110

uric

70 unc

87

70

uric

70unc

90

110

uric

70unc

70 unc

110 SP

140

70unc

70

110

70

uric

110

110

70

90

uric

110

70

110

50

110

110

70unc

sP

110

410

110

70

27

Page 28: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

Guarea oedrata

Guarea thornpsoniiGUIbourtia ehie

Herifjera utilis

Khaya arithothecaKhaya grandifoliolaKhaya ivorensisKlain e doxa ga b o n en SISLophira alataLovoa triohitioides

Mainmea africana

Mansonia 81tissima

Mincia ex Gels airegiaMitra yna sNauclea did errichi

87

Ne s ogo Idonia a a ve rifera

87

Parkia bicolor

uric

Perlcopsis elata

87

Pe te rsia n th us in a oroca rp usPip ta de nia s trum a frica n urn

uric

uric

Pterygota macrocarpa

110

87

Pycn a n th us a rig o1e nsis

110

uric

Rhodognaph alon bre vieuspe

90

87

Stercu/ia rhino etala

70

68

Strom b OSIa 91a u cesce n s

110

uric

Terminal^^ ivorensis

70

110

68

TerminalIa superba

70

410

87

Tieghemella beckeliT

70

58

uric

Triplo chiton s cle fox 10n

70

110

87

APPENDIX 2

Study site and Methodology For Logging StudyThe study was conducted in a 128 ha compartment (RWC4) of the Research Working Circle of Pra-An urnforest Reserve (Figure 4.1; Lat. 6' 159 N and Long. 1' 109 W). The forest covers a total land area of 123 kin'and has undulating topography. It is classified as moist semi-deciduous forest (Hall and Swaine, 1981),where annual precipitation is recorded between 1500 and 1750 min with two maxima, first in May-June andthe second in September-October. The forest experience 5 months dry period with less than 100 mm ofprecipitation.

Timber harvesting in this reserve dates back to 1954 and current harvesting rights are held by TopbellIntegrated company (formerly Amima Timbers) with its processing mill located in Kumasi about 75 kin NorthWest of the forest. Past logging records for the experimental compartment could not be traced, however,there was evidence in the forest of past exploitation as indicated by old skid trails(700 meters) and occasional rotten stumps.

Tu rra e n th us a frica n us

Sources:

110

70

90

68

110

70

uric

uric

110

70

90

uric

70unc

70

110

uric

70

110

90

68

70

110

110

a = Logan (1947); b = FD (1955); c = FD (1972); d = Francois (1989)

uric

70unc

110

90

uric

70

90

uric

uric

110

70

90

uric

70

70

uric

uric

110

70

140

68

70

70

uric

uric

70unc

110

87

uric

110

70

68

uric

70unc

Data collection

Experimental designTwo harvesting intonsities (I Annual Allowable Cut (I PI!, C) = 26.32m' ha~' and 2AAC = 52.63m' ha') withtwo combinations of tree size selection (large and small diameter trees) were imposed in a factorial designFelling intensity treatments were allocated randomly to four sites of 32 hectares each (figure 4.2)

Before selecting trees to be logged, a 100% timber inventory of all commercial timber trees above 50 cm dbh(1.3m above ground) was conducted in each treatment site (400m x 800m) adopting the stock survey andmapping methods prescribed by Ghana forestry department (Forestry Dept, 1995)

With the assumption that felling of large diameter trees creates larger felling gaps while small diameter treescreate smaller gaps, a wider range of diameters would have been included to facilitate planning for theexperimental tree selection. However, permission was not granted by the forestry department for inclusion ofsmaller stems below the minimum-felling diameter (MFD). All trees selected for felling, either large or small

110

68

uric

110

70

uric

70unc

110

90

70 unc

110 SP

uric

70unc

70

110

70unc

110

90

uric

110 90R

90

70

70

70unc

70

110

70

110

50

70

70

70

110 SP

11090R

70

28

Page 29: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

diameter stern, were then, necessarily within the minimum felling diameter for the respective species(Ranging from 50 cm, 70 cm, 90 cm to 110 cm DBH)

The criteria for stern size selection, was therefore to include in the small size category more than 50% ofstems from within 20 cm of the MFD of respective species, In the large size category, more than 50% oftrees were of dbh greater than 20 cm of MFD, Thus, ranges of diameters applied were not as wide asintended for the study. Summary of stem numbers, species numbers, and volume per various combination oflogging treatments is given in table (4.1). For the high intensity treatment it was impossible increasing theharvesting volume without taking more species, given the requirement to select stems only above MFD

With the exception of predetermining the location of primary skid trail, no reduced impact logging methodswere employed. Felling with chainsaw and ground skidding with a Crawler tractor (07 Bulldozer) were themethods used. Conduct of logging crew was riot under the controlled of the research team. Logging tookplace between June and September 2000. Swiss Lumber Company conducted the logging for and on behalfof the concession holder

r~Tablet : Allocation of 10 in treatments to 10 in sites

Treatment s

High intensity (53m ha~ )x Large sterns

Site A

High intensity (53m ha~ )x Small sterns

Site B

Low intensity (23 in3 ha-I)x Small stemsSite C

Low intensity (23 in3 ha-I)x Large sterns

Site D

2.22 Logging damage assessmentTo provide a basis for estimating damage associated with logging, 16 sample plots (40m x loom) wereestablished in each treatment site (representing 2070 of treatment site area 400m x 800m) prior to loggingPlots were locate in stratified random way with a minimum distance of 50m between any two plots.

Within each assessment plot, all trees with stern diameters at breast height (1.3m above ground) equal to orgreater than 50 centimetres were recorded by their species and diameter. Nested subplots were used forsmaller stems; Sub-plot of loin x loom was used for dbh 120cm, loin x 50m for dbh/, O cm, and loin x 25mfor dbh/5cm. All trees in each plot were numbered serial Iy starting with plot number and ending with treenumber (e. g. 0101= tree number I of plot I). Estimates of stocking for each diameter class per plot werecalculated as the actual stems counted multiplied by the proportion of plot size inventoried (i. e. , 50cm dbh x11 20-49cm dbh x 4; I 0-19cm x 8; 5-9cm x 16). Stock estimates from the 16 plots of each treatment site wereused to compute the mean and standard error stocking for the DBH classes at each of the 4 sites. In each ofthe 16 plots in a treatment site, forest canopy opening was assessed with a spherical forest densiometerfrom at least six randomly located points

Logging damage has been expressed by either the number of trees killed or injured for a given density oftrees harvested or measuring basal area of trees damaged or killed for a given basal area of trees harvestedin felling gaps, skid trails and haulage roads (e. g. Gullison and Harder, 1992; Whitman at a1. , I 996)

In this study damage assessment was conducted mainly in the sample plots so as to capture the randomdistribution of logging disturbance. Response variables used for describing logging damage were: forest floorarea disturbed by felling and by skidding; canopy openness; total skid trail length; and number of stems (^5cm dbh) damaged (i. e. injured or killed). To be able to differentiate between felling damage and skiddingdamage, felling assessment was always done before skidding took place in each assessment plot

Ground area disturbed was described as area of forest vegetation disturbed by the fallen tree, skidder andconstruction of log-yards. The longest length of the disturbed floor was measured and perpendicular widthstaken at regular intervals along the length. The area was calculated using the trapezoida! rule,

No. of s ecies

29

22

No. of trees

19

154

19

187

Total vol. in

102

I' "'

L

1600

85

1600

-

.- -

840

^\

..

840

^

.

29

Page 30: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

-J

For additional analysis of skid trail density in treatment sites, distances covered by the skidder were alsosurveyed and mapped. Total treatment area disturbed by the Crawler tractor were calculated as trail lengthmultiplied by average of trail width (taken from at least three locations: generally at the start, inid-section andend of each segment of trail or change of direction). Total Skid trail length and areas (in') in each treatmentsite were determined for the primary, secondary and tertiary trails. Primary trails are defined here as theinidrib of the skid trail network that was used by the skidder on every extraction trip and ran the full length ofthe treatment site. Tertiary trails were those used to extract individual trees. Secondary trails connect primaryto tertiary trails and were used more than once. The log-yard at each treatment site was surveyed (by thetriangulation method) using a trailing tape and prismatic hand~oil compass. The area was determined foreach site as the sum of the triangles composing the yard and differentiated into landings and loading bays.Landings are locations for pilling logs awaiting transportation to mill. Loading bays are areas cleared for thehaulage trucks to pack and get loaded. Portions of loading bays were used as work camps.

Post-harvest canopy openness was taken as the loss in canopy cover resulting from logging activities in theplots. It was determined as the difference between the pre-logging average for each plot and post-loggingopenness (taken from six random locations in each plot) using a spherical forest densiometer

Damage to residual trees were catsgorised as: (1) lightly damaged = when ^25% of tree stem or crown wereaffected; (2) severely damaged = when a 25% parts were affected but still standing and; (3) destroyed =when whole tree was destroyed or killed. For each affected tree, the damage score, stem diameter, species,and initial tree number (where possible) were recorded. All damaged sterns ^5 cm dbh in the 16 sample plotsper treatment were recorded

Implications of minimum felling diameter on residual forest damage were explored on the hypothesis thatdamage to residual trees and vegetation did not differ with sizes of trees felled

Data analysisThe 16 data plots per treatment site were considered as the replicates for each treatment In order todetermine whether total area disturbed differed among treatments, the areas of plots affected were firstcategorised by the logging activities (felling and skidding) for one analysis, and then summed for anotheranalysis. Due to several cases of no disturbance encountered in the plots, treatment effects weredetermined using nonparametric (Kruskal-Wallis test) analysis.

To determine whether total number of sterns damaged differed among treatments, the affected trees per plotwere also Gategorised according to cause of damage (felling or skidding), by diameter (dbh) classes and bydegree of damage (i. e. Injured or killed). The severely damaged and the destroyed categories werecombined as trees killed because of the low occurrence of the severely damaged trees, Variations innumber of stems damaged (by felling and skidding operations) in the various dbh classes among treatmentswere however determined with analysis of variance (ANOVA). Variations in total skid trail among treatmentswere also determined using analysis of variance (ANOVA) in SPSS 11.0. I by Kirinear and Gray (2000),Relationship between number of sterns damaged and area disturbed was examined in a linear regressionanalysis. Differences in these analyses were considered statistical Iy significant at P = 0.05

APPENDIX 3

Methodology for study on wood loss due to logging woundsStudy siteInformation on logging wounds was collected from 3 compartments (RWC1,2.3) of the Research WorkingCircle in Pra-An urn Forest. The three compartments occup approximately 520 ha. They were selectivelylogged between I 995,997 at a mean intensity of 2 trees ha~ . The vegetation is moist semi-deciduous forestwith mean annual rainfall of 1625 min. The predominant timber species with sterns >30 cm dbh in the threecompartments are Geltis mildbraedi7zenkeri, Turfeanthus 8171cana, Storeuffa rhinopetala and Trj710chitonSGIeroxy/on,

Sampling designOn the premise that stern growth increment varies with stem size and species, it was assumed that differentspecies and stem diameters would react differently to wounding during logging, even if environmentalconditions (biotic and abiotic) were uniform throughout the forest

So to investigate the occurrence of decay following wounding, sample stems were taken from four diameterclasses (20-29.30-39.40-49 and 50+ cm dbh) and I5 species in compartments I-3. For each species atleast three samples were taken for each diameter class. Due to low harvest intensity in the study site, it wasdifficult to find many wounded trees within close vicinity. Therefore sample trees were selected in an

.-J

30

Page 31: SOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN · PDF fileSOCIAL AND ECOLOGICAL IMPACTS OF LOGGING IN GHANA ... International Tropical Timber Organization ... (Class I & 2) species during

opportunistic manner along old skid trails and within felling gaps. Due to lack of proper tree climbing gear,observations were made only on wounds located on the lower bote (i. e. 4 meters off ground)

For each sample, species name and stern dbh were recorded. Wounds were recorded by location on stern,size and condition. Location was determined as the above ground distance between the base of tree and theinid-point of the wound area measured in centimetres with a linen tape. Wound size was described as thewound surface area (cm'), calculated as average length multiplied by average width. The length was takenas the longest distance between two opposite ends of the wound. Widths were measured at 2-310cations atright angle to the chosen length. Wound conditions were recorded as I = Healed (completely covered withregenerated bark), and 2 = Unhealed (no regenerated bark and still showing dead wood scar).

It has been observed that external signs of stem injury do not always lead to wood decay beneath theregenerated bark (A1bers and Heyd, 2000). For instance, Trockenbrodt and others (2002) observed thatamong 28 shorea trees identified to have hollow stems, 9 (32%) were riot hollow or had insignificantdeterioration when felled. It was therefore assumed in this study that wounds completely covered withregenerated bark were better protected against deterioration than the uncovered wounds. Therefore, onlywound without regenerated bark were investigated for wood degradation. From more than one point on theuncovered wound surface, stem increment borer was pushed into the wound to detect decay. The decaywas measured as the length (mm) of the wood core showing discoloration or soft wood tissue resulting fromdecaying wound

To estimate potential loss of economic timber due to logging wounds, stem (dbh) basal area of all woundedsample tree were determined and grouped into 4 dbh classes (20-29.30-49,4049 & 50+ cm) and by woundcategories (healed and unhealed). Stem volumes were not used in this study because tree volume equationsderived for Ghanaian timber trees are applicable to stems ^ 30 cm dbh (Wong, I 989)Wounded sample tree included both felling and skid-damaged sterns (e. g. Plate 5.1)

Data analysisResponses to wounding were assessed as the occurrence of healed or unhealed wound and the decaydepth in unhealed wound five years after logging. The factors for comparing responses were species,diameter class of wounded tree, size of wound and wound location

The null hypothesis that the occurrence of healed and unhealed wound conditions on stems followingwounding is independent of species was tested analysis using Chi-square contingency table in SPSS 110.1Correlations between wound location, wound size, decay depth and stem dbh were examined using scattergraphs and Pearson's correlation analysis. Variations in decay depths among species and among sterndiameter classes were examined using two-way analysis of variance (ANOVA) test in SPSS 11.0. I.

r^

L;

,'*

,.~

^-

.

31